Clinical Genetic Testing for Alzheimer Disease and Related Dementias

Dementia is common, affecting an estimated 47 to 55 million people worldwide.^1,2 In the past several decades, genetic contributions to neurodegenerative dementia have been investigated with the goal of improving understanding, treatment, and prevention of these diseases. Although the role of genetics in dementia has not been fully elucidated, current knowledge can be used to improve clinical practice. We provide an overview of the most common hereditary dementias and a framework for genetic testing in a clinical setting.

Causes of Dementia: Monogenic vs Complex

Heritability plays a role in many types of neurodegenerative dementia, but few cases are monogenic or Mendelian (ie, caused by a pathogenic variant in a single disease-associated gene). For example, Alzheimer disease (AD) accounts for about 60% to 80% of all dementia cases, but pathogenic variants confirming monogenic disease are only identified in approximately 1% to 5% of cases.^2,3 The overall molecular diagnostic rate of genetic neurodegenerative disease is likely between 5% and 10%.^2,4 Most monogenic forms of dementia follow an autosomal dominant inheritance pattern in which there is a 50% chance of inheritance for offspring and full siblings of the proband.

By contrast, most dementia cases are considered complex or multifactorial, in which disease is caused by an interplay of multiple genetic, lifestyle, and environmental factors. Complex dementia may also be referred to as “sporadic”, particularly when there is no known family history.⁵ The genetic factors in complex dementia are the result of a multitude of genetic variants with small to moderate effect size rather than a single rare variant with a large effect, as seen in monogenic forms of dementia.^6-8

The discussion of genetic testing in this article is primarily focused on monogenic dementia and the APOE gene, which have the highest utility in a clinical setting. Testing for monogenic dementia can be instrumental in confirming a timely diagnosis and shortening the diagnostic odyssey, particularly when differential diagnoses are difficult to distinguish. Even in cases in which a clinical diagnosis has been confirmed, identifying a monogenic cause of disease can be beneficial in informing inheritance risk for family members and providing additional information about phenotype or progression. Although disease-modifying therapies (DMTs) are not available for many neurodegenerative dementias, genetic testing may provide details that can inform clinical trial eligibility and aid in the development of gene-specific treatment.⁹

Framework for Genetic Testing: Family History and Accurate Phenotyping

Recommendations for the use of genetic testing (Table 1) vary by disease type and are based on available treatment options, frequency of monogenic disease, and reason for testing. Accurate family history collection and phenotyping are essential to ensure testing is provided when a genetic form of dementia is suspected.

Detailed information regarding the proband’s family medical history is necessary to identify patterns resembling autosomal dominant inheritance. Typically, a 3-generation pedigree is collected (Figure). One in 4 individuals aged ≥55 years have a first-degree relative with dementia;⁶ however, most of these cases will be attributed to complex etiology.² Histories correlating with the highest risk of autosomal dominant disease are those with at least 3 affected individuals in 2 generations who are linked by a first-degree relative or individuals with early-onset dementia.^10,11 However, a negative family history does not preclude monogenic dementia. Disease that appears to be sporadic may be attributed to de novo variants or incomplete penetrance. Alternatively, family history may be hidden by limited family structure, misattributed parenthood, early death of relatives by unrelated causes, misdiagnosis, anticipation, or unknown family history.¹¹

Accurate phenotyping of the individual and family is crucial in forming a comprehensive differential diagnosis. The individual should be assessed for motor symptoms, such as tremor, fasciculations, chorea, myoclonus, ataxia, parkinsonism, motor neuron disease, and spasticity, as many types of neurodegenerative dementia are associated with movement disorders.^3,6 Psychiatric features also should be assessed, as they are prevalent in Huntington disease (HD), frontotemporal dementia (FTD), and prion disease. In fact, neuropsychiatric and behavioral changes associated with dementia are often misdiagnosed as psychiatric illness, delaying diagnosis especially in early-onset cases.¹⁰ Early-onset dementia, which accounts for 6% to 9% of cases, is also a red flag for possible genetic etiology.¹² The estimated heritability of early-onset AD is >90%, and 45% of FTD cases are diagnosed between ages 40 and 69 years.^12,13

Types of Genetic Dementia

A brief discussion of phenotype and genetics for the most common genetic dementias is provided in the following section. Additional information regarding indications for genetic testing, associated genes, and testing considerations is provided in Table 1.

Alzheimer Disease

AD, the most common type of dementia, is often categorized into late-onset AD, which accounts for the majority of sporadic cases, and early-onset AD (EOAD), in which symptoms develop before age 65 years.¹ Most cases of monogenic AD, also known as “autosomal dominant AD (ADAD)”, are caused by highly penetrant, pathogenic variants in 1 of 3 genes: PSEN1, PSEN2, or APP.¹⁴ Identification of the causative gene can provide additional information about clinical manifestations and prognosis (Table 2).

The risk of ADAD is highest in individuals with a strong family history of dementia (ie, ≥3 affected family members) or EOAD. The molecular diagnostic yield has been reported to be about 5% to 15% of individuals with EOAD and 60% to 80% of familial EOAD cases.^14,15 Genetic testing for ADAD is appropriate in symptomatic individuals with a strong family history regardless of age of onset, and individuals with age of onset <60 years regardless of family history.^9,11 Further details related to indications for genetic testing are provided in Table 1. Genetic diagnosis can improve the speed and accuracy of an AD diagnosis, which is important for individuals interested in anti-amyloid therapies, which require use in early stages of disease.⁹

Genetic risk factors also contribute to the underlying heritability of complex AD with >30 genetic loci identified in genome-wide studies associated with late-onset AD.^7,8 Genes such as APOE, SORL1, TREM2, and ABCA7 modify the lifetime risk of AD but lack the diagnostic or predictive utility of monogenic AD and are therefore considered only risk factors. However, the testing landscape is always changing. Rare truncating SORL1 variants with high penetrance have been associated with ADAD in the literature, but classification of the SORL1 gene as a monogenic cause of AD is controversial because of limited data on absolute risk and segregation.¹⁶APOE genotyping can be useful in the clinic for reasons other than risk modification as is discussed in detail in the following section. Empiric lifetime risk of AD can also be modified based on family history of AD (Table 2).^3,6

APOE

The APOE gene is the strongest genetic risk factor for AD. APOE has 3 isoforms: APOE2, APOE3, and APOE4, leading to 6 possible genotypes, as each person inherits 1 copy of the gene from each parent (eg, ε4/ε3, ε4/ε4, ε3/ε3).^17,18 Among people of European ancestry, the most common allele is ε3, which is neutral regarding AD risk, whereas the rarer ε2 allele is considered protective and is associated with a later age at onset.^18,19 The ε4 allele is associated with a dose-dependent increased lifetime risk of AD and with an age at onset about 12 years younger than noncarriers.^18,20,21 Risk estimates for each genotype are presented in Table 2.

Genetic risk factors for AD have also been studied in relation to the disparity of AD incidence between ethnic and racial groups. Black and Hispanic/Latino Americans are 2 times and 1.5 times, respectively, more likely than non-Hispanic White (NHW) Americans to develop AD.¹ Although the association of APOE genotype and AD is modulated by ancestry and sex (most notably there are studies suggesting an attenuated risk associated with ε4 for Black Americans as compared to NHW) and ε4 allele frequency varies between racial and ethnic groups, the current understanding of AD genetics does not fully explain the difference in incidence of AD.^22,23 Health disparities produced by social and structural factors play a role in the risk of developing AD.¹

Whereas the role of APOE as a risk modifier for AD has been well-established, this gene has taken on a new significance with the approval of the first antiamyloid therapies. The ε4 allele is associated with an increased risk of amyloid-related imaging abnormalities (ARIAs) during treatment with certain antiamyloid therapies. Results from clinical trials of antiamyloid agents found a dose-dependent relationship between ε4 and the risk of ARIA, with homozygous carriers of ε4 at greatest risk.^24,25 Appropriate-use recommendations for both aducanumab (Aduhelm; Biogen, Cambridge, MA) and lecanemab-irmb (Leqembi; Eisai, Nutley, NJ; Biogen, Cambridge, MA) include APOE genotyping before treatment, and lecanemab’s label includes a black box warning.^26,27 This recommendation is important because the ε4 allele is found at greater frequency among people with AD compared with the general population; 1 trial found that 69% of participants diagnosed with mild cognitive impairment attributable to AD or mild AD had at least 1 ε4 copy, with 16% being ε4 homozygotes.²⁴

Whereas APOE genotyping may be initiated for informed decision-making regarding ARIA risks, genetic testing presents a new challenge to clinicians, who must not only counsel their patients regarding the risks and benefits of antiamyloid therapy, but also lead a discussion about the implications of the individual’s APOE genotype on AD risk for relatives. An incomplete understanding of the role genetic risk factors play in differences between ethnic and racial groups adds additional nuance to counseling of diverse patient populations, and there is a clear need for further research on the topic. Clinicians may need to field inquiries and referrals for presymptomatic APOE genetic testing, which is not recommended for clinical purposes. Traditional pretest and posttest genetic counseling have long been supported as standard of care for any individual considering genetic testing for AD, including APOE genotyping.¹⁰

Cerebral Amyloid Angiopathy

Cerebral amyloid angiopathy (CAA) is related to AD as both include deposition of amyloid-Β. However, they differ in the mechanism of brain injury, with CAA occurring because of cerebrovascular deposition of amyloid-Β, whereas AD is caused by deposition in neuritic plaques. Presenting symptoms vary widely and include cognitive decline, lobar intracerebral hemorrhages, and transient focal neurologic episodes.²⁸ Isolated CAA can occur but is more likely in people with AD, including 45.3% of individuals with severe neuritic plaques as demonstrated in 1 autopsy-based study.²⁹

Genetic testing should be considered in cases of familial, early-onset CAA. Pathogenic variants in the ITM2B gene are known to be causative of familial CAA as are several select APP variants, both of which follow autosomal dominant inheritance patterns. APOE ε4 is also associated with an increased risk of CAA in the context of AD pathology.³⁰

Dementia With Lewy Bodies

Lewy body disease is a spectrum of neurodegenerative conditions including Parkinson disease (PD), PD with dementia (PDD), and dementia with Lewy bodies (DLB).^31-33 DLB, the second most common neurodegenerative dementia after AD,³⁴ has historically been differentiated from PDD by the timing between onset of motor and cognitive symptoms, but biomarkers, CSF analytes, neuroimaging data, genetics, and neuropathology are being studied to improve differentiation among the Lewy body diseases.^31,33-35 DLB is known to have phenotypic and pathologic overlap with PD, PDD, and AD. The rate of cognitive impairment is considered to be faster and risk of delusions is lower in DLB compared with AD. Other symptoms include parkinsonism (especially bradykinesia and rigidity), REM sleep behavior disorder, visual hallucinations, and fluctuating attention.³¹

Few causative genes have been associated with monogenic DLB. Rare pathogenic variants in the SNCA gene can be causative of DLB or PD. Genes associated with increased risk for PD and AD, such as GBA and APOE, may also modify risk of DLB.^31,36 In addition, variants in genes associated with monogenic dementia and PD, such as VPS35, APP, GRN, and PSEN1, have been identified in people with DLB.^34,37,38 Genetic testing is indicated for individuals with a family history of DLB or other dementia, PD, or Gaucher disease.⁹

Frontotemporal Dementia

FTD is a neurodegenerative disease recognized for its effects on personality, social behavior, language, psychiatric symptoms, and, in some cases, motor or neuromuscular involvement. Multiple clinical phenotypes exist, including behavioral variant FTD, 2 types of primary progressive aphasia (semantic and nonfluent variant), progressive supranuclear palsy syndrome, corticobasal syndrome attributable to corticobasal degeneration, and FTD with motor neuron disease.³⁹ These illnesses arise from a family of neuropathologic conditions that preferentially affect the frontal or anterior temporal lobes, which fall under the umbrella term “frontotemporal lobar degeneration”.⁴⁰

Most individuals experience symptom onset in their 50s or 60s, but onset can range between the 20s and 80s.⁴¹ The majority of FTD cases are sporadic, but approximately 40% of affected individuals have a positive family history of FTD. A monogenic cause is identified in about 10% of familial and 5% to 6% of sporadic cases, highlighting the importance of offering testing to all individuals. The majority of pathogenic variants are identified in the C9ORF72, MAPT, and GRN genes, each associated with varied clinical phenotypes.^39,42 A small proportion of cases result from other genes (Table 1). Although clinical trials of investigational treatments are ongoing, there are no proven DMTs available.⁴² All individuals with FTD should be offered a multigene panel including C9ORF72 repeat expansion testing.^9,11

Huntington Disease

HD, the most common monogenic neurodegenerative disease, is characterized by neuropsychiatric symptoms, a progressive movement disorder, and cognitive impairment.⁴³ The quintessential feature of HD is progressive choreiform movements that usually develop in the fourth or fifth decade, but age at onset can vary widely. The course of the disease generally starts with subtle, progressive cognitive changes and psychiatric symptoms.

HD results from a CAG nucleotide repeat expansion in the HTT gene and follows autosomal dominant inheritance. Repeat lengths ≤26 are not associated with disease. Repeats between 36 and 39 are pathogenic with incomplete penetrance, and repeats ≥40 show complete penetrance. An intermediate allele, 27 to 35 repeats, is not typically associated with disease but can increase the risk of HD in later generations. Anticipation, in which higher repeat numbers are associated with earlier age at onset, can occur in successive generations, especially when inherited from the paternal side.⁴⁴ Diagnosis of an affected individual can be confirmed with genetic testing and should follow the Huntington’s Disease Society of America’s protocol.⁴⁵

Leukodystrophies and Leukoencephalopathies

Genetic conditions associated with white matter abnormalities present a diagnostic challenge because of the heterogeneity of phenotypes and causative genes despite often overlapping presentations and radiologic findings. Conditions such as CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy) and CARASIL (cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy) may be mistaken for sporadic vascular dementia, whereas other genetic white matter diseases are difficult to differentiate from inflammatory disease, such as multiple sclerosis. Common features of genetic leukodystrophies and leukoencephalopathies (GLEs) include degeneration of motor control and cognitive function, seizures, vision and hearing disturbances, balance and mobility difficulties, movement disorders, and learning or developmental disabilities.^46,47 A monogenic cause of GLEs is more likely in cases with age at onset <50 years, systemic findings, and positive family history.⁹

Genetic testing is appropriate for individuals with MRI findings of symmetric, confluent leukoencephalopathy for whom acquired causes, such as inflammation, infection, and malignancy, have been ruled out.⁴⁸ Metabolic testing including very-long-chain fatty acids, cholestanol, and plasma amino acids can be diagnostic for many types of metabolic leukodystrophies, and can be ordered before or concurrently with genetic testing. The approach to treatment for most GLEs involves symptom management and surveillance.^9,48 However, for cases in which disease-specific therapy is available, early diagnosis is important (see Table 3).

Prion Disease

The prion diseases—Creutzfeldt-Jacob disease (CJD), fatal familial insomnia (FFI), and Gerstmann-Straussler-Scheinker disease (GSS)—are rare, fatal, neurodegenerative diseases caused by an accumulation of the misfolded prion protein PrPsc.⁴⁹ Age at onset is usually midlife, but can range from the 20s through the 80s.

CJD is the most common prion disease, characterized by rapidly progressive dementia with psychiatric symptoms, behavioral change, language problems, and movement disorders, including parkinsonism, ataxia, and myoclonus. CJD duration is usually 1 to 2 years. The hallmark of FFI is sleep disruption and autonomic problems, ataxia, myoclonus, psychiatric symptoms, and progressive dementia. FFI disease duration is 1 to 3 years. GSS usually begins with gait changes caused by ataxia or parkinsonism, dysarthria, and vision problems, followed by a progressive dementia, with a duration of 2 to 10 years.⁵⁰

Approximately 85% of prion disease cases are sporadic, with 10% to 15% having a genetic etiology (all GSS, most FFI, and 10% to 15% of CJD cases). Fewer than 1% of cases result from iatrogenic sources, including variant CJD caused by bovine spongiform encephalopathy (“mad cow disease”). Genetic cases are attributable to pathogenic variants in the prion protein gene (PRNP) and follow an autosomal dominant inheritance pattern.⁵¹ Diagnosis includes MRI, EEG, and lumbar puncture for tau, 14-3-3 protein, and real-time quaking-induced conversion assay. Complete genetic testing is available through the National Prion Surveillance Laboratory, but some commercial laboratories offer PRNP sequencing.⁵²

Approach to Genetic Testing

Genetic testing for dementia requires a patient-centered approach with an emphasis on preserving autonomy while navigating family dynamics, because obtaining informed consent from individuals with cognitive impairment can raise ethical concerns. Genetic counseling is highly recommended before and after testing for all individuals but is necessary in cases of positive results or predictive testing. It is best practice to test the affected individual before testing unaffected family members.

Testing for asymptomatic individuals at risk for a genetic condition based on family history is known as predictive testing. It is recommended that predictive testing for any monogenic neurodegenerative dementia follow a modified Huntington’s Disease Society of America predictive testing protocol.^10,45

Pretest Discussion

The pretest discussion is often centered on informed consent and the impact genetic testing may have on the individual and their family. There are many benefits of genetic testing, which should be weighed against the possibility of uncertain results, incidental findings, and negative psychosocial effects.² In many cases, diagnosis of monogenic dementia does not affect treatment or availability of preventative care for a progressive and fatal condition, and knowledge of risk to family members can cause additional emotional distress.^3,10 It is important to engage the individual and care partners in a discussion exploring their motivations for testing and possible consequences of the results.

Test Selection

In most instances, the first line of testing is a targeted multigene panel. Because variants in the same gene can result in varied clinical presentations (a phenomenon known as “pleiotropy”), clinical diagnosis and genetic results may be unreliable predictors of one another.¹¹ One study reported that >30% of pathogenic variants identified in individuals with AD were found in genes not typically associated with AD, such as MAPT, GRN, CSF1R, and PRNP.⁵³ Broad panels including genes associated with monogenic AD, DLB, prion disease, and FTD with C9ORF72 repeat expansion testing are recommended because of the clinical and genetic heterogeneity among the conditions.^9,11

If there is high diagnostic certainty or a conservative approach is preferred, panels tailored to 1 clinical syndrome, such as separate tests analyzing only AD or only FTD genes, are available. Whereas testing more genes may increase the chance of identifying a causative variant, it also increases the risk of uncertain results or incidental findings. Single-gene testing is recommended for conditions with characteristic features and low genetic heterogeneity (HD or prion disease), or in situations where there is a known familial monogenic variant. Although there is evidence of certain conditions or founder mutations more frequently found in specific racial or ethnic groups³¹, in general, test selection for monogenic dementia does not need to be tailored or otherwise limited based on these characteristics.

If a panel returns negative or uninformative results but there is strong suspicion of genetic disease, performing more comprehensive testing, such as exome sequencing or genome sequencing, may be appropriate. In addition to testing more genes, many assays have improved accuracy and the capability to detect a wider variety of variant types, such as copy number variants or repeat expansions. Exome sequencing or genome sequencing could be considered a first-line or cost-effective option for conditions such as GLEs when there is high genetic heterogeneity.^9,48

Result Interpretation and Disclosure

Posttest counseling and results disclosure require effective communication and accurate result interpretation.

Uncertain results occur when a variant is identified that cannot be classified as pathogenic or benign with certainty. Whereas it is best practice not to make medical decisions based on an uncertain result, these types of results can be difficult to explain to individuals and may cause anxiety or disappointment.

Positive results may provide a diagnosis in an affected individual, but results with reduced penetrance can cause uncertainty about disease development for a presymptomatic individual or relative. Even when a highly penetrant variant is identified, estimates, but not accurate predictions of age at onset or phenotype, can be provided in conditions with variable presentations.

In the case of a negative result, it is important to communicate residual risk. A negative result from a comprehensive genetic test can indicate a substantially lower risk of monogenic dementia but does not indicate no risk. In addition, genetic test results cannot rule out the possibility of developing dementia of complex etiology later in life. The individual still carries the risks associated with the general population or an elevated empiric risk based on family history.

Challenges to Integrating Genetic Testing Into Practice

One of the greatest challenges to widespread integration of genetic testing into general neurology practice is the availability of both genetic counseling and testing. Physicians can, and do, become proficient in the genetic testing process, but genetic counselors receive specialized training in psychosocial paradigms, informed consent, relaying complex information, risk assessment, and analysis of test results. Genetic counselors are uniquely positioned to address the specialized needs of individuals with a genetic diagnosis and to partner with physicians to navigate the practical aspects of genetic service integration, such as test selection and billing. Patient and physician demand for clinical genetic testing for dementia is accelerating in part because of the recent availability of DMTs. This demand will soon outstrip the capacity of genetics providers, and development of alternative service delivery models is needed to increase efficiency.⁵⁴ The utilization of genetic testing only when clinically indicated ensures the most efficient usage of resources and reduces individual dissatisfaction and confusion.

Individual interest in genetic testing is on the rise as knowledge of the heritability of dementia becomes more common.⁵⁵ However, the cost of genetic testing is a common barrier, as insurance coverage by Medicare and commercial entities is variable. Many individuals seek answers from direct-to-consumer genetic testing offering polygenic risk scores and APOE genotyping, which can lead to misinterpretation of risk and may require discussion in clinic. The increased demand for genetic testing and continuous discovery of new genes requires that providers stay informed about new genetic testing protocols and resources.